Example Note

One line on what this is, in plain words.

Quick summary

• Short bullets.
• The one idea to remember.

Table

Size (sq ft)	Price ($1,000s)
500	100
800	150

Math

The model predicts \(\hat{y}\) from \(x\). For one training example:

\[f_{w,b}(x^{(i)}) = w x^{(i)} + b\]

Diagram

flowchart TB
    A["Training data"] --> B["Train model"] --> C["Predict new input"]

Runnable Python

Tap Run to try it. The first run takes a while because it downloads Python.

import numpy as np

sizes = np.array([500, 800, 1200, 1500, 2000, 2500])
prices = np.array([100, 150, 170, 230, 300, 350])

w, b = np.polyfit(sizes, prices, 1)
print(f"w = {w:.3f}, b = {b:.1f}")
print(f"Predicted price for 750 sq ft: ${w * 750 + b:.0f}K")

This block uses sizes, prices, w and b from the one above, so run that first:

import matplotlib.pyplot as plt

plt.scatter(sizes, prices, marker="x", color="red", label="Training data")
plt.plot(sizes, w * sizes + b, label="Model")
plt.xlabel("Size (sq ft)")
plt.ylabel("Price ($1,000s)")
plt.legend()
plt.show()

A normal, non-runnable code block:

docker restart thm-jekyll-local

Collapsible section

Tap to expand

Term	Meaning
\(x\)	Input
\(y\)	Target

Remember

• One bullet per thing worth remembering.
• Bold only the key term.